JPS60256230A - Digital information transmitting method - Google Patents

Abstract

PURPOSE:To decrease the redundancy when the correcting capability is made identical by constituting >=1 code word in a multiplex code of plural correcting blocks to improve the correcting capability when the same redundancy is used. CONSTITUTION:An information symbol from an input terminal 11 is written in a prescribed address of a memory 12 by using the 2nd memory address generating circuit 18 to decide values of X, Y, Z axes of the memory 12. Then X=Y=Z=1 is fed to the memory 12 via a selector 13 from the 1st memory address generating circuit 14 and D(1, 1, 1) is fetched to a coding/decoding circuit 19. Then a Y address being the output of the circuit 14 is advanced by 1-4, and the X address is advanced sequentially and the result is fed to the circuit 19. A check symbol P is generated from 8 symbols inputted by the circuit 19 based on one equation I. The address of the symbol is designated by the circuit 14 and stored in the position of the memory 12. Then the check symbol is used to improve the correcting capability.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a digital information transmission method for correcting errors in digital information.

[Prior art]

In general, PCM using a magnetic recording/reproducing device (V”rR)
Error correction codes are used in digital audio carriers such as recorders and CDs (compact discs) to reproduce high-quality reproduced sound.

Encoding and decoding of this error correction code are performed after the information is temporarily stored in memory.

FIG. 1 is a memory diagram showing the configuration of a conventional triple code. In the figure, D represents two symbols in the X direction.

Indicates an information symbol area of kl Xk2 d3 symbols in two directions (
・It shows the check symbol area with 1 added, and the total is (nl, kl, dl +1) and (n2. k2.
The minimum distance (dl +i)x (d2
+1)X n4 Xn2 Xn3 with (d3 +1)
It is the sign of the symbol. 1 + dI, nl = 1 + dI,
n2 = 2 + d2. n3 = 3-1d3,
In (a, b, c), a is the code length, b is the number of information symbols, and C
indicates the minimum distance.

The information symbols of kI Xk2 Xk3 symbols generated during the unit time T are once stored in the memory and then encoded. Using the following coordinates, x-xl, y=yI+
2-21” information symbol D(x+, y+, z+),
P check symbol is p (x+, y+, z+), Q check symbol is q (x+, y+, 21), R check j-7
■) Iff of encoding will be explained by expressing the symbol as r (xl, yl, z+).

First, 61 check symbols of the P code are generated from the information symbols in the memory so as to satisfy the formula +11.

However, α is the root of the primitive polynomial f (Xi) of GF(2'). The value is the number of bits of the symbol.

P encoding according to equation (1) is x-1 to k 2 , z -1
3 in ~, and P of k2 Xk3 correction block
Generate a code and exit.

Next, Q code encoding is performed. Q code encoding is performed on information symbols and P check symbols generated by P encoding, and 62 check symbols satisfying equation (2) are generated.

However, equation (2) is an equation for the information symbol region where lly≦, and in the P check symbol region where kl +1≦y≦n1, (1) (x, y, z) in equation (2) becomes p (x, y.

2) will be replaced.

Encoding using equation (2) is 3 for y-1-nl, z=l~
The Q code of the nl Xk3 correction block is generated and the process ends.

In the R code, d31-specific check symbols are added in two directions according to equation (3). However, in the y symbol region of P and Q, D (x, y, z) is p (x, y.

z), q (x, y, z).

Encoding using equation (3) is x=l~n2, )'-1~n
1, the R code of the nI Xn2 correction block is generated, and the encoding is completed.

The information encoded as described above is transmitted as one frame of data consisting of n3 symbols in two directions.

Figure 2 shows an example of the frame format, where S is a synchronization signal, A is additional information, and B is an information symbol and a check symbol consisting of n3 symbols in two directions in Figure 1. The code shown in is transmitted.

On the other hand, on the receiving side, the information symbols and check symbols of the transmitted nl Xnz frame are stored in a memory in the arrangement shown in FIG. 1 and decoded.

First, decoding is the R code syndrome? Determine R using equation (4).

However, u(x, y, z) is the received symbol stored in memory in the arrangement shown in FIG. 1, and is the transmitted symbol plus an error.

In the decoding of R, in the case of gR-τ, it is determined that there is no error,
"0" is set in the error detection flag FR for each block, and "1" is set when S≠O.

In the decoding of Q, using the syndrome SQ shown by equation (5) and the detection flag FR of the R2H code described above, symbols with the flag set are treated as erasures and corrected, and up to 62 erasures ( correct any errors).

If it is corrected, or if it is determined that there is no error with 5Q=O, the Q error detection flag FQ is set to “0” for each block.
If it is determined that correction is not possible, "1" is set.

Next, the decoding of P is similarly performed using equation (6), which is the syndrome SP
and the detection flag FQ of the above Q, and 6
Correct up to one erasure and if corrected or sp -? In this case, the error detection flag FP is set to "0", and in other cases, "1" is sent.

0 The characteristics of this code are not shown in Figure 1.nI Xn2 Xn3
According to this, even if there is an error in the check symbol, there will be no error remaining in the P, Q , H, it is possible to repeatedly decode R-Q-P-R-Q... by using the error detection flag in the previous stage, and it is possible to perform decoding with higher correction ability. can.

Figure 3 shows the parameters of the P and Q codes in Figure 1,
n1 = n2 = 6, k1 = 2 = 4, d1 = d2-2
The memory diagram for z=1 is shown. This code is P,
Both Q and Q are codes with a minimum ratio M3, and it is possible to correct up to two erasures occurring in the X direction or in the X direction. q (5,5
,1) ,q (5,6,1) ,q (6,5,1)
, q (6, 6, 1) is a symbol necessary because both the check symbols of P and Q are included in the codes of P and Q, and is used to correct the check symbol. Although not shown, the relationship between the check symbols of P, R, and Q (!:R is also similar).

Conventional triple codes are configured as described above, so they can be repeatedly decoded and improve correctability, but they have the disadvantage of high redundancy because check symbols are also included in other codes. there were.

In FIG. 3, there are 16 frames of information symbols and 20 frames of check symbols, of which 4 frames are for correcting the check symbols.

[Summary of the invention]

The present invention has been made in order to eliminate the drawbacks of the conventional methods as described above, and includes a digital information transmission method for transmitting information using a multiple code configuration, in which at least one or more of the codes constituting the multiple code is transmitted. By configuring one code word of a code with multiple correction blocks, it is possible to create a code with low redundancy without decreasing the minimum distance, or a code with a large minimum distance without increasing redundancy, that is, with improved correction ability. The purpose of the present invention is to provide a method for transmitting digital information that can be obtained.

[Embodiments of the invention]

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 4 shows an embodiment of the first invention of the present application, in which the redundancy of the P code in FIG. 3 is lowered.

In the P code in Figure 3, two check symbols are added to each of x = 1 to 6, forming a (6,4°3) code in one column, whereas in Figure 4, One check symbol area is provided for each of 1 to 6, and (1
0, 8, 3) constitutes the code.

P (1,5,1), P (2,5,1) is x=1
and x=2 information symbols so as to satisfy equation (7). Also, P (3,5,1), P (4,5,
1> Similarly, X=3. It is generated from x=4 information symbols.

3 Note that the Q and R codes are the same as conventional ones.

As is clear from a comparison between FIG. 3 and FIG. 4, the number of check symbols has been reduced from 20 to 14. That is, the area of 6 frames is reduced and the redundancy is reduced.

Even with this configuration, the minimum distance is 3 in one correction block of the P code, and up to two erasures can be corrected, as in the conventional example shown in FIG. Here, in this embodiment, since the same correction block of the Q code includes two symbols of the same block of the P code, the P code does not hold in the check symbol area of the Q code. Therefore, when iteratively decoding, up to 4 decodings (R-IQ → P-R) have the same correction ability as conventional methods, but if an error occurs in the Q check symbol with further repetitions, It will be less powerful. However, if the method is applicable to a system, it is impossible to repeat it an infinite number of times, and the degree of improvement due to repetition becomes saturated, so there is no problem in practical use.

FIG. 5 shows a block diagram of encoding and decoding,
11 is an input/output terminal for information symbols, 12 is a memory, 19
14 is an encoding/decoding circuit such as a Reed-Solomon code consisting of a multiplier, an adder, and a group of registers; 14 is an X-address generation circuit 15 that determines the value of the X-axis in FIG. 1; and an X-address generation circuit that determines the value of the y-axis. A first memory address circuit 18, which is composed of a circuit 16 and a two-address generation circuit 17 that determines the values of two axes, and specifies a memory address during encoding and decoding.
13 is a second memory address circuit that has the same configuration as the first memory address circuit 14 and specifies a memory address when writing and reading information symbols; 13 is a first memory address circuit 14 and a second memory address circuit; A selector that selects the output from I8 and supplies it to the memory 12 is shown.

Next, first, the operation during recording will be explained.

5. The information symbol inputted from the input terminal 11 is written into a predetermined address of the memory 12 after the values of the X, 5, y, and z axes of the memory 12 are determined by the second memory address circuit 18.

Next, encoding of P code will be explained using FIG. 4 as an example.

x=y=z=l is supplied to the memory 12 by the first memory address circuit 14 via the selector 13, and D (L
l,1) is taken into the encoding/decoding circuit 19.

Furthermore, the output of the X address generation circuit 16 is advanced one by one to D (1, 2, 1), D (1, 3, 1).

D (1, 4, 1') is read out and supplied to the encoding/decoding circuit 19. Next, the output of the X address generation circuit 15 is advanced by one to make X = 2, and the output of the X address generation circuit 16 is advanced from 1 to 4, D (2, 1, 1) to D (2,
4,1> is read out and supplied to the encoding/decoding circuit 19. The encoding/decoding circuit 19 generates check symbols p (1, 5, 1) and p (2, 5, 1) from the input eight information symbols based on equation (7), and performs the check. Each of the symbols is addressed by a first memory address circuit 14 and stored in the memory 12 at the location shown in FIG.

Similarly, p (3,5,1) and p (4,5°1) are generated from four information symbols in the y direction of x=3 and x=4, respectively, and are stored in the memory 12 and z The encoding of P at =1 is completed. Perform this operation z-1, 2,...
. . , all P encoding is completed.

Encoding of the Q code is performed as follows.

That is, four information symbols in each row of FIG.
Based on equation (2) (however, d2 = 2. and 2 = 4.n2
=6) A check symbol is generated and this is stored in memory 1
It is stored in 2.

Similarly to the P code and Q code, the memory address of the R code is controlled by the first memory address circuit 14, and check symbols are added in two directions.

When the encoding of P, Q, and H is completed, the information symbols and check symbols in the memory 12 are sequentially output from the input/output terminal 11 with the memory address controlled by the first memory address circuit 14.

On the other hand, at the time of decoding, the received symbols are temporarily stored in the memory 12 via the input/output terminal []], and the received symbols necessary for the syndrome calculation of the +41151 +61 formula are encoded by controlling the address in the first memory address circuit 14. The data is taken into the decoding circuit 19. The encoding/decoding circuit 19 has a register that stores a detection flag, calculates an erasure error pattern using this detection flag and syndrome, and adds the error pattern to the erroneous received symbol in the memory 12. Corrections will be made accordingly.

When error correction is completed by R→P→Q or repeated decoding, the memory address is controlled by the second memorization circuit 18, and only information symbols are output from the input/output terminal 11.

In this embodiment, when encoding a P-check symbol, one code word is made up of two conventional one-correction blocks, so the redundancy can be reduced while keeping the minimum distance, that is, the same correction ability. It can be reduced to 8 small children. In other words, the correction ability can be improved without increasing the redundancy.

In the above embodiment, the P code is composed of two columns on the same plane, but it may be composed of two or more planes as long as the values of X are different.

FIG. 7 is a diagram for explaining an example in which information symbols on two planes, z-1 and z=2, are used. In FIG. 7, four information symbols lined up in the y direction from x=1 to 4 in z-1 and 41 information symbols lined up in the y direction from x=1 to 4 in z-2.
[Combinations composing the P code of the information symbol of 1a are (z=1.x=1, z-2, x=2), (x-1, x
=2, z-2°x=3>, (z-1, x=3, z=2.
x=4).

(z=l, x=4, z-2, x=1). (
z=I, x=I, z=2. When x=2) is taken as (III), P (1, 5, 1) and P (2, 5, 2) are generated so as to satisfy equation (8).

9 Such an example in FIG. 7 has a remarkable effect in the case of a triple code, etc.

That is,! =1. Assume that an error detection flag is added to the Q code of y=i. At this time, x=l on the 2=1 plane shown in equation (7)! = 2 for the correction block consisting of D (1
, 1, 1) and D (2, 1, 1), but since this P code can correct the erasure of up to 2 symbols, the above error can be corrected.

However, 2=1. If the error detection flag is also set in the Q code of y=2, four symbols will be lost and correction will be impossible. On the other hand, as shown in equation (8), when the l codeword is composed of two planes, D (1, 1,
1), the disappearance of two symbols D (1, 2, 1) is only 1
This does not occur in the correction block, and in this case zero correction is possible.

In this way, when a P code and a Q code are configured on the same plane, the detection flag added to one correction block of Q becomes 2 erasures in the P code, but when the P code is configured on different planes, As a result, only one loss is required, and the correction ability can be improved. Also, when reducing the redundancy of the Q code, the code is constructed across two different value planes so that multiple symbols constituting one correction block of the Q code are not included in one correction block of the P code. , a similar effect can be obtained.

In addition, -! In the first embodiment, an example of a triple code has been described, but the same applies to a double code, a quadruple code, a quintuple code, . . . , an n-code.

FIG. 6 shows the second invention of the present application, which shows an example in which a double code excluding the Q code is applied to the three-dimensionally arranged information symbol block as shown in FIG. There is. In this example, as shown in the figure, (n
3. 3, d3+-1) from the three R correction blocks indicated by the dotted line S.
+1) constitutes a sign. Therefore, compared to the conventional method, the redundancy is the same, the minimum distance is increased, and the correction ability is improved. Furthermore, by configuring the P code in the same manner as the R code, the correction ability can be further improved. Furthermore, if the correction capabilities are made the same, the degree of redundancy can be lowered.

Further, as shown by the dotted line in FIG. 6, the same effect can be obtained even if the code is configured in a broken line across a plurality of blocks.

〔Effect of the invention〕

As described above, according to the present invention, one or more of the codes constituting the conventional multiplex code
Since a code word is made up of multiple correction blocks, the correction ability can be improved if the redundancy is made the same, and the redundancy can be reduced if the correction ability is made the same. be.

[Brief explanation of the drawing]

2 Figure 1 is a memory diagram showing the configuration of a conventional 3m code, and Figure 2 is a frame format for transmitting a conventional 3m code.
- Figure 3 shows the conventional three car row numbers P and Q.
A block diagram of the code, FIG. 4 is a code block diagram according to an embodiment of the first invention of the present application, FIG. 5 is a block diagram of encoding and decoding according to an embodiment of the present invention, and FIG. FIG. 7 is a code configuration diagram when type 2 symbols are used for information symbols in a three-dimensional array, which is an embodiment of the second invention of the present application, and FIG. 7 is a gA of the present application.
1 is a code configuration diagram of another embodiment of the invention of FIG. 12...Memory, 13...Selector, 14...First memory address generation circuit, 18...12 memory address generation circuit, 19...Encoding/decoding circuit. Note that the same reference numerals in the figures indicate the same or equivalent parts. Agent Masuo Oiwa 3 Figure 1 Figure 2 1 Figure 3 23456 Figure 4 23456 Figure 5 4 No. 6 Figure 7 Continued from page 1 0 Inventor Ishi 1) Noriyoshi Sadada Nagaokakyo City Horse M Research Institute 144-

Claims (1)

[Claims] fll + (+=1 to L, L≧2) in each i-dimensional direction
For an information symbol block consisting of pols, in the i-th dimension code (CI code) for one symbol (information symbol string), each (rz- +) number (+) to be added to the information symbol string is e) nl: number of symbols in one block of c1 code) check (j
When < i, ej=n+, when j>i, e1=k J
) In the digital information transmission method, the encoding process for creating additional blocks of III is performed for all l from 1 to shima, and information is transmitted as L-fold codes of the above-mentioned number of symbols. ..., one or more of the L) codes constitutes l codewords each consisting of m (≧2) correction blocks consisting of an information symbol string and a check symbol string, and A digital information transmission method characterized in that a check symbol of one code word made up of m correction blocks is equally allocated to an information symbol string of each of m correction blocks. (2) The m correction blocks constituting one code word of the CI code are all on the same plane of the planes formed by this dimension and another dimension. A digital information transmission method according to claim 1, characterized in that: (3) Each of the m correction blocks constituting one code word of the above Ct code is on a different plane among the planes formed by this dimension and another dimension. A digital information transmission method according to claim 1, characterized in that: (4) kl in each i-dimensional direction (+=1 to L, I, ≧2)
For an information symbol block consisting of Pol, in the i-th dimension code (Ci code) that adds λj to k 1IIIJ symbols (information symbol string) on the back, each (+ to rz-) to be added to the information symbol string. For any i from 1 to 1, the encoding process creates a 3 = k ) additional blocks consisting of (nl: c + number of symbols in one block of code) check symbols (check symbol string). In a digital information transmission method in which information is transmitted as a multiplex code using the entire information symbol and check symbol, )1 CJ (IE(i
l) One or more of the codes is a correction pro-7i m(
≧2) each constitutes one code word, and L notation m
1. A digital information transmission method characterized in that check symbols of l codewords made up of m correction blocks are equally allocated to information symbol strings of m correction blocks.